*Sky & Telescope Illustration by Casey Reed
Privacy Policy | Contact Us | Affiliates | About Us | Participant Sites
International Year of Astronomy | American Association of Variable Star Observers (AAVSO)
Privacy Policy | Contact Us | Affiliates | About Us | Participant Sites
International Year of Astronomy | American Association of Variable Star Observers (AAVSO)
Hello Heinz,I'm working on an AIP4WIN tutorial. I currently use the attached spreadsheet for my final computations. It's very rough now but I was planning to include something like this in my submission to Brian.Your suggestions would be appreciated.Thanks.Tom
I think calling this "very rough" is an understatement :-), it's already quite cool IMHO.I think for an "end-user" Excel sheet, there's actually not that much that is missing.1) fool-proofing: Highlight the cells that the user has to enter data into, and lock the cells that contain data/formulas that should not be overwritten by accident (comparison star date, calculated fields)2) allow a wider choice of comparison stars (lambda Aur in your case). Allow to use more than one comp star.3) maybe unify the TC calculating section and the measurement section What do you think?
Hello again Heinz,These are great ideas. I'll work on each of them and check back with you with the result - hopefully soon.Tom
That's exactly what I was thinking!I haven't done any DSLR reduction so I don't know what the equations are (I assume it's just differential photometry). If the group can come up with the equations, I could see if I could assemble it.I assume someone in this group already has something like this already done. Anyone care to share?
The complexity of this depends greatly on whether we are ambitious enough to consider extinction exactly (figuring out the airmass for each star separately etc), or not (just assume identical extinction across the field).
I guess for the first version we can do with the 2nd option?
In that case it boils down to the usual rather simple system of equations:
M_measured_var = M_var_raw + TC * B_minus_V_var + Z M_ci = M_ci_raw + TC * B_minus_V_ci + Z (i=1 ..n for n comparison stars)
where we know: M_raw_* : raw instr. magnitudes B_minus_V_* : color indices of the stars M_ci : "real" magnitudes of comparison stars
and we don't know
TC : color coefficient of Z : Zero point at time of exposure (so let's assume we can figure in extinction here as well as we assume it's approx the same across the field...as 1st order approx.). and of course M_measured_var the measurement of the magnitude
My dream Excel Sheet would cover the following cases automagically:
Case 1: Only one comparison star ==> the system of equations is underdetermined :-), so the user has to supply the value TC himself. Or he/she has to do at least one measurement with many comparison stars (see following cases ) and use the TC for the next sessions.
Case 2: Two comparison stars (e.g. eta and zeta Aur) We have a set of two linear equations in two unknown vars=> we can actually solve this analytically. We probably want to check that the system is well-conditioned (I think it will be iff the color indices of the comp stars are not too close together). So here the user would have the option to use either * a previously evaluated TC and eliminate Z by subtracting the two equations we've got or * use the TC and Z that would make the two comparison stars come out with exactly the right magnitudes to get the variable's mag.
Case 3: More than two comparison stars: The system is overdetermined, so we would get a TC by least-squares fit (linear regression). Optionally the user would also be able to force a previously evaluated TC again, in that case he would get a M_measured_var for each comparison star.
Well, computationally cases 2 and 3 are the same , linear regression with two data points IS solving a linear 2 x 2 system and the error is always 0 :-).
I think the neat thing of such an Excel sheet would be the option to do either a TC approximation or a measurement on a fixed TC. EDIT: The layout could be like in the attachment, but after seeing Tom's sheet I think it's a must to include some diagrams as well for the TC fitting and for the fitting of the individual measurements
CU Heinz-Bernd